Text Localization in Video Corpus: A Comprehensive Survey

Text information extraction from natural scene videos is a rising area of research. Since text in natural scene videos generally carries valuable details, detecting and recognizing scene text has been deemed essential for a variety of advanced computer vision applications. There has been a lot of effort put into extracting text regions from scene text videos in an effective and reliable manner. As most text recognition applications have high demand of robust algorithms for detecting and localizing texts from a given scene text video, so the researchers mainly focus on the two important stages text detection and text localization. Nowadays detection of textual information from videos are very challenging and exciting research area in video processing and machine learning field. Text detection finds a vital role in current applications such as indexing, easy and efficient retrieval, keyword-based image search and event identification. However, the text region detection from video has several challenges like low resolution, complex background, alignment of text and variation in size, color, style. The ample of works have been done on text detection, and all these considered different properties to distinguish the text region from its background in a video frame. In this paper, a comprehensive survey on text localization from video and image frames are discussed, identifies research gaps and highlighted the future directions.